PSI - Issue 54

Mihaela Iordachescu et al. / Procedia Structural Integrity 54 (2024) 52–58 Mihaela Iordachescu et. al / Structural Integrity Procedia 00 (2023) 000–000

57

6

the steel given in Table 2. The two curves incorporated into the diagram correspond to the SSY limit condition (determined by the limit value K = 31 MPa·m ½ ) and to the tensile load F Y resulting from the combined tension and bending mechanism of plastic collapse indicated in Fig. 4:

A f A 0

⎞ ⎠⎟ 2

A f A 0

F Y A 0

A f A 0

F F 0

⎛ ⎝⎜

(4)

=

= 1 − 2

+ 2

−

The experimental results shown in Fig. 4 are couples of tensile load and crack depth values occurring simultaneously and corresponding to: a) fracture in air of fatigue pre-cracked SENT specimens (red triangles); b) fracture in air of SENT specimens previously cracked in the SSRT-FIP tests (red dots), c) fracture of fatigue pre cracked SENT specimens when subjected to the SSRT-FIP testing (white dots) and d) distinct steps of the environmentally assisted cracking process produced in 3 of the SSRT-FIP tests, two of which ended with the specimens failed in FIP medium, while the third specimen was broken in air, after interrupting the SSRT-FIP test. The crack depth of d) was calculated from the applied load and COD with Eq (1) for values below a ssy ; for higher values it was visually measured on the fracture surface or on the sequence of lateral images acquired with VIC-2D during the tests.

Fig. 4. Diagram of damage tolerance with results of the carried out tests

The points of Fig. 4 representing the failure combinations of load and crack depth (critical cracking) roughly locate on or slightly beyond the curve that predicts the plastic collapse of the resistant ligament, both in FIP medium and in air, and regardless of whether this critical crack initiated and grew by fatigue or by hydrogen uptake during slow strain rate tensile loading. The only effect of assisted cracking on failure seems to be the reduction of the margin by which failure load exceeds the theoretical plastic collapse load. This effect is particularly remarkable when subcritical cracking is fatigue produced. The experimental results are also of interest when assessed in terms of the stress intensity factor, since the hydrogen-assisted cracking threshold in FIP medium is lower than the limit value of the SSY regime (31 MPa m½) of the tested specimens. Assisted cracking hardly initiates before this regime stops, but just after being surpassed crack growth is strongly accelerated with no load increase until the resistant ligament becomes plastically collapsed